def test_add_impact_severity(variant):
plugin = VcfPlugin()
variant.most_severe_consequence = 'stop_gained'
plugin._add_impact_severity(variant)
assert variant.impact_severity == 'HIGH'
def test_add_impact_severity(variant):
plugin = VcfPlugin()
variant.most_severe_consequence = 'stop_gained'
plugin._add_impact_severity(variant)
assert variant.impact_severity == 'HIGH'
def test_add_most_severe(variant):
plugin = VcfPlugin()
variant.consequences = ['downstream_gene_variant', 'stop_gained']
plugin._add_most_severe_consequence(variant)
assert variant.most_severe_consequence == 'stop_gained'
def test_add_most_severe(variant):
plugin = VcfPlugin()
variant.consequences = ['downstream_gene_variant', 'stop_gained']
plugin._add_most_severe_consequence(variant)
assert variant.most_severe_consequence == 'stop_gained'
def test_individual(case_obj):
adapter=VcfPlugin()
adapter.add_case(case_obj)
ind_id = "ADM1059A1"
ind_obj = adapter.individual(ind_id)
assert ind_obj.ind_id == ind_id
def test_filters_no_filter(self, case_obj):
plugin = VcfPlugin()
plugin.add_case(case_obj)
case_id = case_obj.case_id
filters = {}
result = plugin.variants(case_id, filters=filters, count=1000)
assert result.nr_of_variants == 108
def test_filters_no_filter(self, case_obj):
plugin = VcfPlugin()
plugin.add_case(case_obj)
case_id = case_obj.case_id
filters = {}
result = plugin.variants(case_id, filters=filters, count=1000)
assert result.nr_of_variants == 108
def test_variant():
vcf_plugin = VcfPlugin(root_path=vcf)
variant = vcf_plugin.variant('hapmap.vcf', 'X_155239821_G_A')
assert variant['CHROM'] == 'X'
assert variant['POS'] == '155239821'
# get 10th variant
variant = vcf_plugin.variant('hapmap.vcf', '3_124998098_C_A')
assert variant['index'] == 10
def test_add_compounds(variant):
plugin = VcfPlugin()
info_dict = {
'Compounds': "643594:11_120999413_G_A>24|11_120988038_T_G>23|11_120991571_G_A>10|11_121028581_C_G>36"
}
plugin._add_compounds(variant, info_dict)
compounds = variant.compounds
highest_scored_compound = compounds[0]
assert int(highest_scored_compound.combined_score) == 36
def test_format_variants(cyvcf_variant, case_obj, header):
vcf_plugin = VcfPlugin()
vcf_plugin.head = header
vcf_plugin.vep_header = header.vep_columns
vcf_plugin.snpeff_header = header.snpeff_columns
variant_obj = vcf_plugin._format_variants(cyvcf_variant, index=1,
case_obj=case_obj, add_all_info=False)
assert variant_obj.CHROM == cyvcf_variant.CHROM
assert variant_obj.start == cyvcf_variant.start
def test_variant(case_obj):
vcf_plugin = VcfPlugin()
vcf_plugin.add_case(case_obj)
variant = vcf_plugin.variant(case_obj.case_id, 'X_155239821_G_A')
assert variant['CHROM'] == 'X'
assert int(variant['POS']) == 155239821
# get 10th variant
variant = vcf_plugin.variant(case_obj.case_id, '3_124998098_C_A')
assert variant['index'] == 10
def load(ctx, variant_source, family_file, family_type):
"""
Load a case into the database.
This can be done with a config file or from command line.
If no database was found run puzzle init first.
"""
db_path = ctx.obj['db_path']
if not os.path.exists(db_path):
logger.warn("database not initialized, run 'puzzle init'")
ctx.abort()
logger.debug('Set puzzle backend to {0}'.format(ctx.obj['mode']))
mode = ctx.obj['mode']
logger.debug('Set puzzle mode to {0}'.format(ctx.obj['variant_type']))
variant_type = ctx.obj['variant_type']
if mode == 'vcf':
logger.info("Initialzing VCF plugin")
if not family_file:
logger.error("Please provide a ped like file")
ctx.abort()
try:
plugin = VcfPlugin(
root_path=variant_source,
case_lines=family_file,
case_type=family_type,
vtype=variant_type
)
except SyntaxError as e:
logger.error(e.message)
ctx.abort()
elif mode == 'gemini':
logger.debug("Initialzing GEMINI plugin")
try:
plugin = GeminiPlugin(db=variant_source, vtype=variant_type)
except NameError:
logger.error("Need to have gemini installed to use gemini plugin")
ctx.abort()
except DatabaseError as e:
logger.error("{0} is not a valid gemini db".format(variant_source))
logger.info("variant-source has to point to a gemini database")
ctx.abort()
logger.debug("Plugin setup was succesfull")
# from gemini can create multiple cases
store = SqlStore(db_path)
for case_obj in plugin.cases():
# extract case information
logger.debug("adding case: {}".format(case_obj['case_id']))
store.add_case(case_obj, vtype=variant_type, mode=mode)
def test_variants_case_no_ped():
vcf_plugin = VcfPlugin(root_path=vcf)
variants = vcf_plugin.variants(case_id='hapmap.vcf')
variant = next(variants)
assert variant['CHROM'] == 'X'
assert int(variant['POS']) == 84563218
assert variant['index'] == 1
variant = next(variants)
assert variant['CHROM'] == '2'
assert int(variant['POS']) == 233349186
assert variant['index'] == 2
def test_variants_case_with_ped():
vcf_plugin = VcfPlugin(root_path=vcf, case_lines=individuals, case_type='ped')
variants = vcf_plugin.variants(case_id='636808')
variant = next(variants)
assert variant['CHROM'] == 'X'
assert int(variant['POS']) == 84563218
assert variant['index'] == 1
variant = next(variants)
assert variant['CHROM'] == '2'
assert int(variant['POS']) == 233349186
assert variant['index'] == 2
def test_filters_frequency(self, case_obj):
plugin = VcfPlugin()
plugin.add_case(case_obj)
case_id = case_obj.case_id
filters = {'frequency':'0.001'}
result = plugin.variants(case_id, filters=filters, count=1000)
variants = result.variants
nr_of_variants = result.nr_of_variants
for variant_obj in variants:
assert variant_obj.max_freq <= 0.001
assert nr_of_variants == 73
def test_filters_frequency(self, case_obj):
plugin = VcfPlugin()
plugin.add_case(case_obj)
case_id = case_obj.case_id
filters = {'frequency':'0.001'}
result = plugin.variants(case_id, filters=filters, count=1000)
variants = result.variants
nr_of_variants = result.nr_of_variants
for variant_obj in variants:
assert variant_obj.max_freq <= 0.001
assert nr_of_variants == 88
def test_filters_impact_severities_high(self, case_obj):
plugin = VcfPlugin()
plugin.add_case(case_obj)
case_id = case_obj.case_id
filters = {'impact_severities':['HIGH']}
result = plugin.variants(case_id, filters=filters, count=1000)
variants = result.variants
nr_of_variants = result.nr_of_variants
for variant_obj in variants:
assert variant_obj.impact_severity == 'HIGH'
assert nr_of_variants == 7
def select_plugin(case_obj):
"""Select and initialize the correct plugin for the case."""
if case_obj.variant_mode == 'vcf':
logger.debug("Using vcf plugin")
plugin = VcfPlugin(root_path=case_obj.variant_source,
vtype=case_obj.variant_type)
plugin.case_objs = [case_obj]
elif case_obj.variant_mode == 'gemini':
logger.debug("Using gemini plugin")
plugin = GeminiPlugin(db=case_obj.variant_source,
vtype=case_obj.variant_type)
case_id = case_obj.case_id
return plugin, case_id
def test_filters_gene_ids(self, case_obj):
plugin = VcfPlugin()
plugin.add_case(case_obj)
case_id = case_obj.case_id
filters = {'gene_ids':['POF1B']}
result = plugin.variants(case_id, filters=filters, count=1000)
variants = result.variants
nr_of_variants = result.nr_of_variants
for variant_obj in variants:
assert 'POF1B' in variant_obj.gene_symbols
assert nr_of_variants == 1
def test_filters_cadd(self, case_obj):
plugin = VcfPlugin()
plugin.add_case(case_obj)
case_id = case_obj.case_id
filters = {'cadd':'20'}
result = plugin.variants(case_id, filters=filters, count=1000)
variants = result.variants
nr_of_variants = result.nr_of_variants
for variant_obj in variants:
variant_obj.cadd_score >= 20
assert nr_of_variants == 50
def test_filters_consequence(self, case_obj):
plugin = VcfPlugin()
plugin.add_case(case_obj)
case_id = case_obj.case_id
filters = {'consequence':['frameshift_variant']}
result = plugin.variants(case_id, filters=filters, count=1000)
variants = result.variants
nr_of_variants = result.nr_of_variants
for variant_obj in variants:
assert 'frameshift_variant' in variant_obj.consequences
assert nr_of_variants == 4
def test_filters_impact_severities_high(self, case_obj):
plugin = VcfPlugin()
plugin.add_case(case_obj)
case_id = case_obj.case_id
filters = {'impact_severities':['HIGH']}
result = plugin.variants(case_id, filters=filters, count=1000)
variants = result.variants
nr_of_variants = result.nr_of_variants
for variant_obj in variants:
assert variant_obj.impact_severity == 'HIGH'
assert nr_of_variants == 7
def test_filters_cadd(self, case_obj):
plugin = VcfPlugin()
plugin.add_case(case_obj)
case_id = case_obj.case_id
filters = {'cadd':'20'}
result = plugin.variants(case_id, filters=filters, count=1000)
variants = result.variants
nr_of_variants = result.nr_of_variants
for variant_obj in variants:
variant_obj.cadd_score >= 20
assert nr_of_variants == 50
def test_filters_consequence(self, case_obj):
plugin = VcfPlugin()
plugin.add_case(case_obj)
case_id = case_obj.case_id
filters = {'consequence':['frameshift_variant']}
result = plugin.variants(case_id, filters=filters, count=1000)
variants = result.variants
nr_of_variants = result.nr_of_variants
for variant_obj in variants:
assert 'frameshift_variant' in variant_obj.consequences
assert nr_of_variants == 4
def test_filters_gene_ids(self, case_obj):
plugin = VcfPlugin()
plugin.add_case(case_obj)
case_id = case_obj.case_id
filters = {'gene_ids':['POF1B']}
result = plugin.variants(case_id, filters=filters, count=1000)
variants = result.variants
nr_of_variants = result.nr_of_variants
for variant_obj in variants:
assert 'POF1B' in variant_obj.gene_symbols
assert nr_of_variants == 1
def test_variants_case(case_obj):
vcf_plugin = VcfPlugin()
vcf_plugin.add_case(case_obj)
#case_id is 'hapmap.vcf' since no ped file is given
result = vcf_plugin.variants(case_id=case_obj.case_id)
variants = result.variants
variant = variants[0]
assert variant['CHROM'] == 'X'
assert int(variant['POS']) == 84563218
assert variant['index'] == 1
variant = variants[1]
assert variant['CHROM'] == '2'
assert int(variant['POS']) == 233349186
assert variant['index'] == 2
def test_variants_case(case_obj):
vcf_plugin = VcfPlugin()
vcf_plugin.add_case(case_obj)
#case_id is 'hapmap.vcf' since no ped file is given
result = vcf_plugin.variants(case_id=case_obj.case_id)
variants = result.variants
variant = variants[0]
assert variant['CHROM'] == 'X'
assert int(variant['POS']) == 84563218
assert variant['index'] == 1
variant = variants[1]
assert variant['CHROM'] == '2'
assert int(variant['POS']) == 233349186
assert variant['index'] == 2
def test_add_consequences(variant):
plugin = VcfPlugin()
# The raw variant line is treated like a string so it does not matter
# if it is malformed
raw_variant_line = "7\t98994144\t.\tA\tC\t51.68\tPASS\tAC=1;AF=0.167;AN=6"\
";BaseQRankSum=-1.965;CADD=13.51;CSQ=C|downstream_gene_variant|MODIFIER|"\
"ARPC1B|ENSG00000130429|Transcript|ENST00000491294|retained_intron|||||||"\
"||||1724|1|HGNC|704||||||||||||||,C|downstream_gene_variant|MODIFIER|"\
"ARPC1B|ENSG00000130429|Transcript|ENST00000451682|protein_coding||||||||"\
"|||1745|1|HGNC|704||CCDS5661.1|ENSP00000389631|ARC1B_HUMAN|F8VXW2_HUMAN"\
"&C9K057_HUMAN&C9JTT6_HUMAN&C9JQM8_HUMAN&C9JM51_HUMAN&C9JFG9_HUMAN&C9JEY1"\
"_HUMAN&C9JBJ7_HUMAN&C9J6C8_HUMAN&C9J4Z7_HUMAN&A4D275_HUMAN|UPI0000125D33|"
plugin._add_consequences(variant, raw_variant_line)
assert variant.consequences == ["downstream_gene_variant"]
def select_plugin(self, case_obj):
"""Select and initialize the correct plugin for the case."""
if case_obj.variant_mode == 'vcf':
logger.debug("Using vcf plugin")
plugin = VcfPlugin(case_obj.variant_type)
elif case_obj.variant_mode == 'gemini':
logger.debug("Using gemini plugin")
plugin = GeminiPlugin(case_obj.variant_type)
#Add case to plugin
plugin.add_case(case_obj)
self.variant_type = case_obj.variant_type
case_id = case_obj.case_id
return plugin, case_id
def select_plugin(self, case_obj):
"""Select and initialize the correct plugin for the case."""
if case_obj.variant_mode == "vcf":
logger.debug("Using vcf plugin")
plugin = VcfPlugin(case_obj.variant_type)
elif case_obj.variant_mode == "gemini":
logger.debug("Using gemini plugin")
plugin = GeminiPlugin(case_obj.variant_type)
# Add case to plugin
plugin.add_case(case_obj)
self.variant_type = case_obj.variant_type
case_id = case_obj.case_id
return plugin, case_id
def test_add_consequences(variant):
plugin = VcfPlugin()
# The raw variant line is treated like a string so it does not matter
# if it is malformed
raw_variant_line = "7\t98994144\t.\tA\tC\t51.68\tPASS\tAC=1;AF=0.167;AN=6"\
";BaseQRankSum=-1.965;CADD=13.51;CSQ=C|downstream_gene_variant|MODIFIER|"\
"ARPC1B|ENSG00000130429|Transcript|ENST00000491294|retained_intron|||||||"\
"||||1724|1|HGNC|704||||||||||||||,C|downstream_gene_variant|MODIFIER|"\
"ARPC1B|ENSG00000130429|Transcript|ENST00000451682|protein_coding||||||||"\
"|||1745|1|HGNC|704||CCDS5661.1|ENSP00000389631|ARC1B_HUMAN|F8VXW2_HUMAN"\
"&C9K057_HUMAN&C9JTT6_HUMAN&C9JQM8_HUMAN&C9JM51_HUMAN&C9JFG9_HUMAN&C9JEY1"\
"_HUMAN&C9JBJ7_HUMAN&C9J6C8_HUMAN&C9J4Z7_HUMAN&A4D275_HUMAN|UPI0000125D33|"
plugin._add_consequences(variant, raw_variant_line)
assert variant.consequences == ["downstream_gene_variant"]
def test_setup_no_file():
"""Test to initialize a gemini database without any file"""
adapter = VcfPlugin()
assert adapter.variant_type == 'snv'
assert adapter.filters.can_filter_gene == True
assert adapter.filters.can_filter_inheritance == True
def test_filters_range(self, case_obj, indexed_vcf_file):
plugin = VcfPlugin()
case_obj.variant_source = indexed_vcf_file
case_obj.compressed = True
case_obj.tabix_index = True
plugin.add_case(case_obj)
case_id = case_obj.case_id
start = 1771120
end = 1771130
filters = {'range':{'chromosome':'1', 'start':start, 'end':end}}
result = plugin.variants(case_id, filters=filters, count=1000)
variants = result.variants
nr_of_variants = result.nr_of_variants
for variant_obj in variants:
assert variant_obj.start >= start
assert variant_obj.stop <= end
assert nr_of_variants == 1
def test_filters_range(self, case_obj, indexed_vcf_file):
plugin = VcfPlugin()
case_obj.variant_source = indexed_vcf_file
case_obj.compressed = True
case_obj.tabix_index = True
plugin.add_case(case_obj)
case_id = case_obj.case_id
start = 1771120
end = 1771130
filters = {'range':{'chromosome':'1', 'start':start, 'end':end}}
result = plugin.variants(case_id, filters=filters, count=1000)
variants = result.variants
nr_of_variants = result.nr_of_variants
for variant_obj in variants:
assert variant_obj.start >= start
assert variant_obj.stop <= end
assert nr_of_variants == 1
class BaseConfig:
PROJECT = PROJECT_NAME
DEBUG = False
TESTING = False
# http://flask.pocoo.org/docs/quickstart/#sessions
SECRET_KEY = 'secret key'
PUZZLE_BACKEND = VcfPlugin()
STORE_ENABLED = False
# default blueprints
BLUEPRINTS = [public_bp, variants_bp]
#Phemonizer credentials
PHENOMIZER_AUTH = False
def test_variant(case_obj):
vcf_plugin = VcfPlugin()
vcf_plugin.add_case(case_obj)
variant = vcf_plugin.variant(case_obj.case_id, 'X_155239821_G_A')
assert variant['CHROM'] == 'X'
assert int(variant['POS']) == 155239821
# get 10th variant
variant = vcf_plugin.variant(case_obj.case_id, '3_124998098_C_A')
assert variant['index'] == 10
def test_format_variants(cyvcf_variant, case_obj, header):
vcf_plugin = VcfPlugin()
vcf_plugin.head = header
vcf_plugin.vep_header = header.vep_columns
vcf_plugin.snpeff_header = header.snpeff_columns
variant_obj = vcf_plugin._format_variants(cyvcf_variant, index=1,
case_obj=case_obj, add_all_info=False)
assert variant_obj.CHROM == cyvcf_variant.CHROM
assert variant_obj.start == cyvcf_variant.start
def test_add_genetic_models_no_models(variant):
plugin = VcfPlugin()
info_dict = {}
plugin._add_genetic_models(variant, info_dict)
assert variant.genetic_models == []
def test_individuals(case_obj):
adapter=VcfPlugin()
adapter.add_case(case_obj)
individuals = [ind for ind in adapter.individuals()]
assert len(individuals) == 3
def test_add_most_severe_no_info(variant):
plugin = VcfPlugin()
plugin._add_most_severe_consequence(variant)
assert variant.most_severe_consequence == None
def vcf():
db = VcfPlugin()
return db
def test_add_genetic_models(variant):
plugin = VcfPlugin()
info_dict = {'GeneticModels': '643594:AD_dn|AR_comp_dn'}
plugin._add_genetic_models(variant, info_dict)
assert set(variant.genetic_models) == set(['AD_dn', 'AR_comp_dn'])
def test_setup_variant_type():
"""docstring for test_setup_variant_type"""
adapter = VcfPlugin('sv')
assert adapter.variant_type == 'sv'
def test__add_individual(individual):
adapter=VcfPlugin()
adapter._add_individual(individual)
assert len(adapter.individual_objs) == 1
def test_add_rank_score_no_score(variant):
plugin = VcfPlugin()
info_dict = {}
plugin._add_rank_score(variant, info_dict)
assert variant.rank_score == None
def test_add_most_severe_no_info(variant):
plugin = VcfPlugin()
plugin._add_most_severe_consequence(variant)
assert variant.most_severe_consequence == None
def vcf():
db = VcfPlugin(TestConfig.PUZZLE_VCF_FILE)
return db
def test_add_compounds_no_compounds(variant):
plugin = VcfPlugin()
info_dict = {}
plugin._add_compounds(variant, info_dict)
assert variant.compounds == []
def test_add_cadd_score(variant):
plugin = VcfPlugin()
info_dict = {'CADD': '24'}
plugin._add_cadd_score(variant, info_dict)
assert float(variant.cadd_score) == 24.0
def load(ctx, variant_source, family_file, family_type, root, mode,
variant_type):
"""
Load a variant source into the database.
If no database was found run puzzle init first.
1. VCF: If a vcf file is used it can be loaded with a ped file
2. GEMINI: Ped information will be retreived from the gemini db
"""
if root is None:
#This is the default puzzle folder
root = os.path.expanduser("~/.puzzle")
if os.path.isfile(root):
logger.error("'root' can't be a file")
ctx.abort()
logger.info("Root directory is: {}".format(root))
db_path = os.path.join(root, 'puzzle_db.sqlite3')
logger.info("db path is: {}".format(db_path))
if not os.path.exists(db_path):
logger.warn("database not initialized, run 'puzzle init'")
ctx.abort()
logger.debug('Set puzzle backend to {0}'.format(mode))
logger.debug('Set variant type to {0}'.format(variant_type))
if mode == 'vcf':
logger.info("Initialzing VCF plugin")
try:
plugin = VcfPlugin(
root_path=variant_source,
case_lines=family_file,
case_type=family_type,
vtype=variant_type
)
except SyntaxError as e:
logger.error(e.message)
ctx.abort()
elif mode == 'gemini':
logger.debug("Initialzing GEMINI plugin")
try:
plugin = GeminiPlugin(db=variant_source, vtype=variant_type)
except NameError:
logger.error("Need to have gemini installed to use gemini plugin")
ctx.abort()
except DatabaseError as e:
logger.error("{} is not a valid gemini db".format(variant_source))
logger.info("variant-source has to point to a gemini database")
ctx.abort()
logger.debug("Plugin setup was succesfull")
# from gemini can create multiple cases
store = SqlStore(db_path)
for case_obj in plugin.cases():
if store.case(case_id=case_obj.case_id).case_id == case_obj.case_id:
logger.warn("{} already exists in the database"
.format(case_obj.case_id))
continue
# extract case information
logger.debug("adding case: {}".format(case_obj.case_id))
store.add_case(case_obj, vtype=variant_type, mode=mode)
def test_add_rank_score(variant):
plugin = VcfPlugin()
info_dict = {'RankScore': '643594:24'}
plugin._add_rank_score(variant, info_dict)
assert int(variant.rank_score) == 24